DESCRIPTION: (Applicant's Abstract) The primary aim of this proposed 4 year study compares the effects of an Enhanced Cognitive-Behavioral (E-CB) Intervention to a "Standard Care" Information Condition in reducing HIV risk among culturally diverse, poor minority male adolescent offenders (N=440). Since the E-CB intervention is offered in the context of court-mandated treatment, it provides a unique opportunity to reach an extremely high risk group that might otherwise be impossible to engage in an intense and specialized HIV risk reduction intervention. Although some of the drug treatment program activities share common components to the E-CB, these activities do not a) incorporate the motivationally, enhanced E-CB format of encouraging participants to actively assist in developing their own risk-reduction strategic to best suit their circumstances; b) include HIV risk specific mixed gender groups with facilitated interactions focused on rehearsing strategies to reduce risk; and c) emphasis on cognitive-behavioral strategies to cope with the transition from in-patient status to community living. AIM 2 is to evaluate if E-CB reduces drug use to a greater extent than SC. AIM 3 is to evaluate if drug abuse (particularly crack cocaine) may mediate the relationships between our interventions and HIV risk outcome. AIM 4 is to evaluate if key variables mediate the relationships between our interventions and HIV risk and drug abuse outcome. AIM 5 is to evaluate if key variables may moderate the relationships between our interventions and our HIV risk and drug abuse outcome. An EXPLORATORY AIM is to evaluate if E-CB produces greater positive therapeutic impact than SC on HIV testing behavior and results. Risk behavior, drug use, HIV and STD status, HIV testing acceptance, drug and other clinically significant phenomena will be measured at pre- and at 6 months post- intervention.. Multiple methods (in-depth interviewing, STD and drug urine analyses) will be used to corroborate and supplement self reports. Repeated measures MANOVA and regression analyses will be used to analyze the effects of E-CB on HIV risk mediating and outcome variables and to explore the relationships between mediating, moderating and outcome variables.